The five-year survival rate for patients with distant colorectal cancer metastasis is 10%. Retinoids have been shown to decrease cancer cell growth, but their role in metastasis has received little attention. Our long-term objective is to understand how dietary vitamin A (retinol) inhibits colorectal cancer progression. We have shown that retinol decreases retinoic acid-resistant colon cancer cell invasion, and Akt phosphorylation, beta-catenin, and matrixmetalloproteinase 7 (MMP7) levels in vitro. Our aims are (1) to determine if vitamin A inhibits metastasis in vivo and (2) to elucidate the signaling pathway retinol uses to inhibit invasion. Aim 1 will use nude mice injected intrasplenically with human colon cancer cells to induce liver metastases. Twenty mice will consume a vitamin A sufficient control diet. Additional mice will consume a diets containing increasing amounts of vitamin A up to 200,000 ILJ/kg diet. Hepatic metastases number and size will be determined. We expect vitamin A supplementation to decrease metastases number and size. Aim 2 will determine the signaling pathway by which retinol decreases invasion. Retinol decreases Akt phosphorylation and constitutively active Akt blocks the effect of retinol on invasion. Decreased Akt phosphorylation results in increased GSKSbeta activity. GSKSbeta targets beta-catenin for proteosomal degradation. Excess nuclear beta-catenin increases the expression of genes favoring metastasis (e.g.MMP7). Constitutively active and dominant negative forms of Akt, and overexpression or siRNA technology to alter GSKSbeta and beta- catenin levels will determine the effect of each protein on cell invasion, growth, and MMP7 mRNA levels. To link Aim 1 and 2 each protein will be measured in tumors excised from the mice used in Aim 1. We expect that retinol will decrease Akt and GSKSbeta phosphorylation as well as beta-catenin, and MMP7 levels in vitro and in vivo. The goals of this application are consistent with those of the NCI and PA# 04-108. Specifically, we will examine the ability of a nutrient, vitamin A, to prevent colon cancer metastasis and determine the signaling pathway retinol uses to decrease metastasis. The survival rate from colon cancer metastasis is very low. Vitamin A decreases the ability of colon cancer cells to move and invade tissues in laboratory experiments. This application will determine if vitamin A can decrease metastasis in mice and the proteins affected by vitamin A that regulate metastasis.